Synthesis and Reactivity of 2-Iodophosphinines - Organometallics

Mohammed Ghalib , Peter G. Jones , Sergej Lysenko , and Joachim W. Heinicke. Organometallics 2014 33 (3), 804-816. Abstract | Full Text HTML | PDF | P...
1 downloads 0 Views 294KB Size
Download PDF
794

Organometallics 1996, 15, 794-801

Synthesis and Reactivity of 2-Iodophosphinines Herman T. Teunissen and Friedrich Bickelhaupt* Scheikundig Laboratorium, Vrije Universiteit, De Boelelaan 1083, 1081 HV Amsterdam, The Netherlands Received July 11, 1995X

The synthesis of 2-iodophosphinines 5 (a, parent compound; b, 4,5-dimethyl derivative) and of their pentacarbonyltungsten complexes 7 is described starting from dichloro(diiodomethyl)phosphine (2), which was prepared via alkylation of phosphorus trichloride with (diiodomethyl)magnesium chloride (1). The halogen-metal exchange reactions of 5a/b and 7a/b are reported: the lithiation reactions of 5a/b with n-BuLi were not successful. In contrast, the complexes 7a/b were converted to the corresponding lithium derivatives 11a/b with n-BuLi in THF at -100 °C. Derivatization of 11a/b with triphenyltin chloride furnished 13a/b, respectively. The reaction of 5b with magnesium in THF at room temperature led to the formation of 3,4-dimethylphosphinine (18) (41.5%), 4,4′,5,5′-tetramethyl-2,2′-biphosphinine (16) (1.3%), and 2-(iodomagnesio)-4,5-dimethylphosphinine (17) (19.2%); 37% of the starting material had decomposed. In a reaction with zinc, 5b gave the organozinc derivatives 20 or 21 in THF/TMEDA or DMF, respectively. The same zinc insertion procedures were applied to 7b but did not furnish the pentacarbonyltungsten complexes 22 or 24 in a clean fashion. However, 22 could be prepared via complexation of 20 with (acetonitrile)pentacarbonyltungsten (6). Introduction The synthesis of phosphinine derivatives, which was first reported by Ma¨rkl in 1966,1 is usually achieved by a multistep procedure in which the phosphinine is prepared from non-phosphaaromatic prcursors.2 In principle, preparation of a new phosphinine via chemical modification of a halophosphinine might be an interesting approach to obtain a variety of functionalized phosphinines. Along this line, Mathey et al. investigated the chemistry of 2-chloro- and 2-bromophosphinines.3 As we were interested in phosphinines functionalized at the 2-position, we focused our attention on the synthesis and reactivity of 2-iodo derivatives in the hope that the enhanced reactivity of the carbon-iodine bond would open new perspectives for investigations on 2-halophosphinines in general. In preliminary communications4 we reported on the synthesis of 2-iodophosphinines and on several exploratory results concerning their reactivity. In this paper we wish to report the synthesis of 2-iodophosphinines 5a/b and their pentacarbonyltungsten complexes 7a/b with full experi-

mental details. Furthermore, metalation reactions of these compounds with n-butyllithium, magnesium, and zinc are described. While halogen-metal exchange with organolithium reagents and the preparation of Grignard reagents is well-known, direct zinc insertion into C-X bonds is less common. However, recently, Knochel provided some important contributions to the preparation of organozinc compounds.5 One of these5a concerns the preparation of arylzinc iodides in DMF or DMAC (dimethylacetamide) via direct zinc insertion into an aryl-iodine bond. Another zinc insertion into an sp2-hybridized C-X bond was reported by Jiang and Xu.6 With 2-bromotrifluoropropene as the substrate, they investigated a variety of solvents (Et2O, THF, DME, and DMF) and a zinc/ silver couple in the presence or absence of TMEDA. It was found that in the presence of TMEDA (1 equiv) in THF, DME, or Et2O, the corresponding organozinc derivative was obtained in high yield while poor results were obtained in DMF. Results and Discussion

* To whom correspondence should be addressed. E-mail: bicklhpt@ chem.vu.nl. X Abstract published in Advance ACS Abstracts, December 15, 1995. (1) Ma¨rkl, G. Angew. Chem. 1966, 78, 907. (2) Ma¨rkl, G. In Multiple Bonds and Low Coordination in Phosphorus Chemistry; Regitz, M., Scherer, O. J., Eds.; Thieme Verlag: Stuttgart, Germany, 1990; pp 220-257. (3) (a) Le Floch, P.; Mathey, F. Tetrahedron Lett. 1989, 30, 817. (b) Le Floch, P.; Ricard, L.; Mathey, F. Polyhedron 1990, 9, 991. (c) Le Floch, P.; Carmichael, D.; Mathey, F. Organometallics 1991, 10, 2432. (d) Le Floch, P.; Carmichael, D.; Mathey, F. Bull. Soc. Chim. Fr. 1992, 129, 291. (e) Le Floch, P.; Carmichael, D.; Mathey, F. Phosphorus Sulfur 1993, 76, 33. (f) Le Floch, P.; Ricard, L.; Mathey, F. J. Chem. Soc., Chem. Commun. 1993, 789. (g) Le Floch, P.; Carmichael, D.; Ricard, L.; Mathey, F. J. Am. Chem. Soc. 1993, 115, 10665. (h) Le Floch, P.; Kolb, A.; Mathey, F. J. Chem. Soc., Chem. Commun. 1994, 2065. (i) Trauner, H; Le Floch, P.; Lefour, J. M.; Ricard, L.; Mathey, F. Synthesis 1995, 717. (4) (a) Teunissen, H. T.; Bickelhaupt, F. Tetrahedron Lett. 1992, 33, 3537. (b) Teunissen, H. T.; Bickelhaupt, F. Bull. Soc. Chim. Belg. 1992, 101, 609. (c) Bickelhaupt, F. Pure Appl. Chem. 1993, 65, 621. (d) Teunissen, H. T.; Bickelhaupt, F. Phosphorus Sulfur 1993, 76, 75.

0276-7333/96/2315-0794$12.00/0

Iodophosphinines. The synthesis of 2-iodophosphinines and their pentacarbonyltungsten complexes was achieved by the Diels-Alder approach developed by Mathey for the corresponding chlorine and bromine derivatives3a and is outlined in Scheme 1. The precursor 2 was prepared by alkylation of phosphorus trichloride with the magnesium carbenoid 1. The freshly prepared solution of 17 was frozen in liquid nitrogen at -196 °C. Then a Schlenk tube containing (5) (a) Majid, T. N.; Knochel, P. Tetrahedron Lett. 1990, 31, 4413. (b) Berk, S. C.; Yeh, M. C. P.; Jeong, N.; Knochel, P. Organometallics 1990, 9, 3053. (c) Rozema, M. J.; Sidduri, A. R.; Knochel, P. J. Org. Chem. 1992, 57, 1956. (d) Tucker, C. E.; Majid, T. N.; Knochel, P. J. Am. Chem. Soc. 1992, 114, 3983. (6) Jiang, B.; Xu, Y. J. Org. Chem. 1991, 56, 7336. (7) Seyferth, D.; Lambert, R. L. J. Organomet. Chem. 1973, 54, 123.

© 1996 American Chemical Society

2-Iodophosphinines

Organometallics, Vol. 15, No. 2, 1996 795 Scheme 1

Scheme 2

Scheme 3

freshly distilled, frozen (-196 °C) phosphorus trichloride (13 equiv) was attached up side down on top of the frozen solution of 1. The Schlenk tube was allowed to warm such that phosphorus trichloride was dropping at its melting point (-112 °C) down on top of the frozen solution of 1. After the phosphorus trichloride had been added, the cooling bath was removed and the reaction mixture warmed to -100 °C. At this temperature, the mixture began to melt and the reaction could occur in the presence of a large excess of phosphorus trichloride, which was expected to suppress di- and trialkylation. In spite of its successful synthesis from 1, 2 (δ(31P) 155.2 ppm) could not be purified by distillation or crystallization. Therefore, the subsequent Diels-Alder reactions with a 1,3-butadiene were carried out with freshly prepared crude 2, which, according to 31P NMR spectroscopy, was approximately 70% pure. The Diels-Alder reactions were carried out at relatively low temperatures: -5 °C for 1,3-butadiene (a series, 23 equiv) and 0 °C for 2,3-dimethylbutadiene (b series, 20 equiv). A large excess of triethylamine (a, 11 equiv; b, 17 equiv) was added to a solution of 2 in the diene to generate the unstable phosphaalkene 3 and to aromatize its primary Diels-Alder adducts 4; to achieve this, the reaction mixture was allowed to warm slowly to room temperature. The aromatization of 4 was monitored by 31P NMR spectroscopy. The aromatization of 4a required less than 16 h. On the other hand, 55 h stirring at room temperature was necessary to convert 4b (δ(31P) 94 ppm) to 5b. The 2-iodophosphinines were isolated by distillation under high vacuum in a yield of 32% (5b, yellow crystals (δ(31P) 216 ppm)) and 10% (5a, yellow liquid (δ(31P) 233 ppm)) relative to iodoform, the precursor of 1 (Scheme 1).7 Whereas 5b was isolated in pure form, 5a was slightly contaminated with 1,1-diiodoisobutane (5-10%) which is formed as a byproduct, presumably in the preparation of 1 from iodoform and isopropylmagnesium chloride (Scheme 1). Distillative separation of 1,1-diiodoisobutane was easy for 5b but impossible in the case of 5a. Phosphinines 5 were fully characterized by NMR, MS, and elemental analysis. The NMR assignments were supported by 2D NMR techniques. While HH COSY, CH COSY, COLOC, and 1D NOE did not distinguish between C4 and C5 in 5b, INADEQUATE solved this problem conveniently. The 1J(CC) values of the phosphinine ring lie in the range 55.3-58.7 Hz, which is

close to the value of benzene8 (57.0 Hz) and confirms the electronic delocalization within the aromatic ring. Furthermore, the 1J(CC) values involving the methyl groups (42.6-43.67 Hz) are very close to those in o-xylene8 (44.2 Hz). The pentacarbonyltungsten complexes 7a (δ(31P) 202 ppm, 1J(PW) ) 286 Hz) and 7b (δ(31P) 185 ppm, 1J(PW) ) 280.4 Hz) were prepared in moderate to high yield by reacting 5a,b with (acetonitrile)pentacarbonyltungsten9 (6) in THF at room temperature (7a) or 45-50 °C (7b) (Scheme 1). In these reactions, a small excess of 6 (about 10%) was necessary to drive the complexation to completion. In the reactivity studies, the attention was focused on the dimethyl derivatives 5b and 7b, as 5b could be obtained in pure form and in higher yield. Furthermore, its synthesis is less complicated than that of 5a because of the easy handling of 2,3-dimethyl-1,3-butadiene compared with 1,3-butadiene. Attempts to prepare the η6 complex 9 by heating 5b for several hours under reflux in di-n-butyl ether with hexacarbonylchromium analogous to the procedures of No¨th10 failed; besides decomposition products of 5b, only the η1 complex 8 (δ(31P) 231 ppm) was formed (Scheme 2). Apart from decomposition, the lack of steric hindrance at positions 2 and 6 of the phosphinine ring probably is an important factor for this failure.11 Compound 8 was isolated in low yield (12%) and was fully characterized by NMR and MS spectroscopy. Organolithium Derivatives. Toward 1.6 M nBuLi, there are some interesting differences in behavior between uncoordinated 2-iodophosphinines 5 and their pentacarbonyltungsten complexes 7. As outlined in Scheme 3, the lithiation of uncoordinated 2-iodophosphinines did not occur, whereas the pentacarbonyltungsten complexes showed the expected iodine-lithium exchange. Reaction of 1.6 M n-BuLi (1 equiv) with 5a/b in THF was performed at -100 °C. After 0.5 h of (8) Wray, V. Prog. Nucl. Magn. Reson. Spectrosc. 1979, 13, 177. (9) Marinetti, A.; Bauer, S.; Ricard, L.; Mathey, F. Organometallics 1990, 9, 793. (10) Deberitz, J.; No¨th, H. Chem. Ber. 1970, 103, 2541. (11) Nainan, K. C.; Sears, C. T. J. Organomet. Chem. 1978, 148, C31.

796

Organometallics, Vol. 15, No. 2, 1996

stirring of the black reaction mixtures at temperature below -80 °C, a quench reaction with ClSnR′3 (R′ ) Ph (5b), R′ ) CH3 (5a)) was carried out. After warming of the samples to room temperature, 31P NMR spectroscopy indicated complete decomposition in the case of 5b and the formation of many non-phosphaaromatic compounds with a poor signal to noise ratio in the case of 5a. In the latter case, the experiment was not conclusive with respect to the lithiation of 5a, because of the presence of 1,1-diiodoisobutane. The lithiation of the pentacarbonyltungsten complexes was successful, and the organolithium phosphinines 11a/b could be detected by low-temperature 31P NMR spectroscopy. The thermal decomposition of 11a/b was followed as a function of temperature; 11b (δ(31P) 163.4 ppm, 1J(PW) ) 239.8 Hz) turned out to decompose at -70 °C. In the 31P NMR spectra, signals appeared between 0 and -100 ppm and around 240 ppm. At room temperature the most important signals were located at 242 and 7 ppm with a rather unsatisfactory signal to noise ratio. The signal at 242 ppm did not show tungsten satellites so this compound might be the Fischer carbene 12 (Scheme 3), which could be formed by an intramolecular attack of the C-Li bond on CO. Such intramolecular reaction would result in a strained four-membered ring system, which opens by disruption of the coordination between phosphorus and tungsten; it is reasonable to assume that the resulting vacant coordination site at tungsten is filled by a Lewis base such as THF or CO to furnish an 18 electron species as indicated. The spectroscopic detection of 11a was more difficult as its formation was not selective and the thermal stability rather low. The first recorded spectrum at -85 °C showed the presence of 11a (δ(31P) 176.9 ppm, 1J(PW) ) 235.2 Hz) and many signals in the region of 8 to -52 ppm. The spectrum recorded at -80 °C showed already substantial decomposition, thus indicating that methyl groups at the 4- and 5-positions lead to increased stability of (organometallic) phosphinine derivatives. Both organolithium derivatives were quenched with triphenyltin chloride at -90 or -100 °C, and the organotin derivatives 13a/b were isolated in pure form and were fully characterized. The yield of 13b was moderate (48%), but 13a was obtained in a rather disappointing yield (4%). The steric shielding of the 2-position of the phosphinine by the W(CO)5 moiety might be responsible for a rather slow reaction of triphenyltin chloride. The competition between decomposition of 11a/b on the one hand and nucleophilic substitution at tin on the other is more detrimental for 11a than for 11b. Mathey et al. encountered similar unexpected difficulties in quench reactions of a molybdenum-coordinated 2-lithiophosphinine.3d Apparently, substituents can have a substantial influence on the effectiveness with which an organolithiophosphinine is derivatized. In addition, the purification of 13a turned out to be difficult; many crystallizations were required before it was isolated in pure form. The question why halogen-lithium exchange is not successful in the case of 5a/b, whereas for the pentacarbonyltungsten complexes, notably 7b, it is achieved rather easily, is difficult to answer. Mathey et al. suggested that a strong stabilizing interaction between the phosphaaromatic system and the halogen (Cl, Br) could account for the inertness of free 2-halophos-

Teunissen and Bickelhaupt Scheme 4

phinines3b toward lithiation at the 2-position;3c it was suggested that this stabilization is less pronounced in the pentacarbonyltungsten complex in which the aromaticity of the phosphinine ring is somewhat reduced.12 We propose that, in 2-iodophosphinines 5a/b, reaction with the organolithium reagents results in a competition between nucleophilic attack at phosphorus and halogenmetal exchange. Due to the interaction between the 2-halogen substituent and the phosphinine ring, which seems operational in 2-iodophosphinines, too, the reaction is dominated by nucleophilic attack at phosphorus to give 14 (Scheme 4).13 Possibly, decomposition proceeds via cleavage of the C-I bond, which leads to the formation of 15, a phosphinin-2-ylidene derivative which has been shown to be a local minimum by ab initio calculations.14 Its polymerization may be the cause of decomposition; Scheme 4 shows only this decomposition pathway, but others are conceivable. Apparently, nucleophilic attack at phosphorus is less competitive in 7, probably for steric reasons. In conclusion, the halogen-lithium exchange of neither 5a/b nor the 2-bromo- or 2-chlorophosphinines3b could be achieved. In contrast, the pentacarbonyltungsten complexes 7a/b are easily converted to the corresponding 2-lithiophosphinines 11a/b by n-BuLi. The pentacarbonyltungsten complex of a 2-bromophosphinine has been reported to undergo halogen-metal exchange in reaction with phenyllithium;3c so in this respect our 2-iodophosphinines do not exhibit unique behavior. Organomagnesium Derivatives. In contrast to the iodine-lithium exchange, reaction of 5b with magnesium was slightly more successful in leading to the formation of three products, along with substantial decomposition (Scheme 5). The best results were obtained in THF, using three times sublimed magnesium rather than commercial magnesium turnings. The magnesium was first activated with 1,2-dibromoethane; after cooling of the reaction mixture to room temperature, 5b was added and the reaction was followed by 31P NMR spectroscopy. Two of the products were known: 3,4-dimethylphosphinine (18, δ(31P) 191 ppm)15 and 4,4′,5,5′-tetramethyl-2,2′-biphosphinine (16, δ(31P) 181.3 ppm).16 This compound was unambiguously identified by its proton-coupled 31P NMR spectrum which showed an AA′BB′XX′ system. With the PH (12) Alcaraz, J. M.; Mathey, F. Tetrahedron Lett. 1984, 25, 207. (13) Reference 2, p 239. (14) Nyulaszi, L; Szieberth, D; Veszpremi, T. J. Org. Chem. 1995, 60, 1647. (15) Alcaraz, J. M.; Mathey, F. Tetrahedron Lett. 1984, 25, 4659. (16) Le Floch, P.; Carmichael, D.; Ricard, L.; Mathey, F. J. Am. Chem. Soc. 1991, 113, 667.

2-Iodophosphinines

Organometallics, Vol. 15, No. 2, 1996 797 Scheme 5

Scheme 6

Table 1. Relative Intensities of 31P NMR Signals during the Reaction of 5b with Magnesium in THF time (h)

5b

18

16

17

2.5 5.75 21.0 25.0

6.50 0.34 0 0

1 1 1 1

0 0 0.16 0.20a

1.83 1.57 0.59 1.04a

a Due to the relatively large line width of 18, these figures are not a good indication for relative amounts; actually, these amounts are smaller.

coupling constants given by Mathey for 16,16 simulation of the proton-coupled 31P NMR spectrum exactly reproduced the experimentally obtained pattern. The third product was the desired phosphinine Grignard reagent 17 (δ(31P) 229.0 ppm). This compound was identified by a reaction with triphenyltin chloride which gave the organotin derivative 19 (δ(31P) 222.0 ppm), identical with a product obtained via an independent route.4a Several observations may be of interest. First, it is remarkable that hardly any reaction occurred with diethyl ether instead of THF as solvent. Also, the degree of decomposition of 5b increased when activation of magnesium with 1,2-dibromoethane was carried out in the presence of 5b. Finally, when magnesium salts, formed by activation of magnesium with 1,2-dibromoethane, were removed by washing with THF, the reaction on subsequent addition of 5b was very slow. This observation may indicate an initiating17 or catalytic18 role of magnesium bromide in the reaction of magnesium with 5b. By calibration with triphenylphosphine as an internal standard, the absolute yields of 18, 16, and 17 were found to be 41.5, 1.3, and 19.2%, respectively; more than one-third (37%) of 5b had decomposed. The reaction of 5b with magnesium in the presence of triphenylphosphine turned out to be considerably slower (see Experimental Section) than in the absence of triphenylphosphine. A reaction mixture consisting of magnesium (4.11 mmol) in THF (5 mL) was activated with 1,2-dibromoethane (2.32 mmol). After addition of 5b (1.16 mmol), the reaction was followed by 31P NMR spectroscopy at room temperature during 25 h. The course of the reaction is presented in Table 1. From Table 1 it follows that the conversion of 5b is complete within 21 h. Furthermore, it is remarkable that the formation of 16 does not occur in the initial phase of the reaction but only after the amount of 5b has diminished substantially and approaches zero. (17) Kharasch, M. S.; Reinmuth, O. Grignard Reactions of Non Metallic Substances; Prentice-Hall: New York, 1954; p 56. (18) Garst, J. F.; Lawrence, K. E.; Batlow, R.; Boone, J. R. Inorg. Chim. Acta 1994, 222, 365.

Finally, the Grignard reagent 17 decomposed slowly at room temperature. The course of the reactions discussed above are challenging from a mechanistic point of view. The formation of 18 and 17 can be explained by the radical mechanism of Grignard reagent formation.17-19 The obvious possibility that 16 might be formed by a coupling reaction between 5b and 17 is ruled out by the observation that 16 is not formed in the initial phase when 5b and 17 are abundant. The alternative formation of 16 by nucleophilic attack of 17 on 18 is unlikely because nucleophilic attack on the phosphinine ring occurs at the phosphorus center.13 More likely is the assumption that 17 decomposes (possibly under the influence of magnesium bromide and the excess of magnesium) under the formation of the 2-phosphininyl radical, which will yield 16 by radical coupling and 18 by hydrogen abstraction from the solvent. Mathey et al. described that 2-chloro-4,5-dimethylphosphinine did not show any reaction with magnesium in refluxing THF under ultrasonic irradiation;3b the 2-bromo derivative proved to be equally inert. A reinvestigation showed that, in reaction of 2-bromo-4,5dimethylphosphinine with magnesium alone, the phosphinine Grignard reagent could not be generated efficiently.3d,e On the other hand, trapping of the in situ generated Grignard species with silicon electrophiles under Barbier conditions lead to the formation of functionalized phosphinines in good yields.3d,e Therefore, the reaction of 5b with magnesium is the first example of the direct reaction leading to an observable phosphinine Grignard reagent (17); this is possibly a consequence of the high reactivity of the carbon-iodine bond. However, interesting as the formation of 17 and that of 16 from 5b may be from a mechanistic point of view, their yield is too low for preparative applications. Organozinc Derivatives. The course of the metalation of 5b with zinc in THF/TMEDA or DMF is depicted in Scheme 6. In THF, the selective formation of 20 (δ(31P) 220.5 ppm) occurred at room temperature by reaction of 5b with zinc dust (1.8 equiv) in the (19) (a) Walborsky, H. M. Acc. Chem. Res. 1990, 23, 286. (b) Bickelhaupt, F. J. Organomet. Chem. 1994, 475, 1.

798

Organometallics, Vol. 15, No. 2, 1996

presence of TMEDA (1 equiv). The reaction required 20 h, after which the conversion of 5b to 20 (and traces of 18) was complete. The NMR spectroscopic characterization of 20 was performed on crystals obtained after cooling its THF solution for several days to -70 °C. Without TMEDA, zinc insertion of 5b did not occur. Even prolonged heating of 5b in THF at 60 °C with zinc dust resulted only in slow reduction of 5b to 18. In DMF, the conversion of 5b to 21 (δ(31P) 222.6 ppm) and traces of 18 was achieved by reaction with zinc dust (2.8 equiv) during 20 h at room temperature. The organometallic phosphinine derivative turned out to be very stable; a sealed tube containing 21 in DMF-d7 (prepared for NMR characterization) remained practically unchanged after several months at room temperature. These results on zinc insertion reactions of 5b are unique, as attempted insertion reactions with the 2-chloro and 2-bromo derivatives were unsuccessful.3b Comparison of the zinc insertions into iodobenzene and 5b shows the latter to be more reactive; iodobenzene is converted to phenylzinc iodide for only 80% during 22 h at 55 °C in DMF.5a As the zinc insertion in THF/ TMEDA has been described for an olefinic system only,6 a direct comparison is not possible. Therefore, we carried out a zinc insertion with iodobenzene in THF/ TMEDA at room temperature and quenched the reaction mixture with bromine; according to GCMS analysis, zinc insertion had not occurred. A possible explanation for these differences in reactivity between 5b and iodobenzene might be the increased electron acceptor properties of the phosphinine ring compared with those of the benzene ring.20 As the rate-determining step in the zinc insertion reaction is presumably a single electron transfer from zinc to the arene ring,5a 5b would be expected to react faster than iodobenzene. Whereas the zinc insertion into 5b is successful in DMF and THF/TMEDA, the corresponding tungsten complex 7b showed a disappointing behavior under the same reaction conditions. The desired zinc complexes 22 and 24 were not formed selectively; the 31P NMR spectra showed many signals in the region 0-200 ppm. A different approach for the preparation of 22 and 24 is the reaction of 20 or 21 with (acetonitrile)pentacarbonyltungsten 6 (Scheme 6). This strategy failed in DMF; 24 was not obtained, but slow decomposition of 21 occurred. In THF/TMEDA, however, 22 (δ(31P) 174.5 ppm, 1J(PW) ) 249.7 Hz) was formed from 20 (containing a small amount of 18) in moderate yield. The complexation procedure at room temperature required 20 to 30 h stirring when carried out with 1 equiv of 6 (yield 68-70%). With an excess (13%) of 6, the complexation was complete within 15 h, and the yield was slightly higher (78%). Other phosphinines present in the final reaction mixture were 7b (5.6%), 18 (4.3%), and 23 (3.5%). The formation of 22 was supported by 1H and 13C NMR spectra which were recorded in THFd8. Unfortunately, 22 decomposed slowly at room temperature. Following this process by NMR showed that, after 7 days, 22 was no longer present; 23 and two unidentified products (δ(31P) 60.8 ppm, 1J(PW) ) 260.4 Hz; 61.5 ppm, 1J(PW) ) 259.5 Hz) were formed. (20) Elschenbroich, C.; Nowotny, M.; Metz, B.; Massa, W.; Graulich, J. Angew. Chem. 1991, 103, 601.

Teunissen and Bickelhaupt

Conclusions The synthesis of 2-iodophosphinines and their pentacarbonyltungsten complexes is an interesting expansion of the chemistry of halophosphinines. The unique properties of the 2-iodophosphinines were illustrated by the synthesis of organozinc derivatives 20-22, which cannot be obtained from the 2-bromo or 2-chloro analogues. Experimental Section General Procedures. All oxygen- and/or water-sensitive reactions were carried out under dry nitrogen with oven-dried glassware and oxygen-free, dry solvents. THF was distilled first from NaH and finally from Na/benzophenone. Pentane was distilled from LiAlH4, and CH2Cl2 was distilled from P2O5. Triethylamine was distilled from CaH2, and PCl3 was distilled before use. A solution of 1.6 M n-BuLi in hexane was commercially available and used as received. Sublimed magnesium was kindly provided by G. Schat. 2-Chloropropane and 2,3-dimethyl-1,3-butadiene were distilled before use. TMEDA was distilled from Na. DMF was distilled and stored on 4 Å molecular sieves. Zinc powder was activated with dilute HCl and thoroughly washed with oxygen-free water, ethanol, THF, and pentane. NMR spectra were recorded at a Bruker AC 200 spectrometer at 200 MHz (1H) or 50.32 MHz (13C). For 31P NMR spectroscopy a Bruker WM 250 was used operating at 100.26 MHz. Incidentally, 1H, 13C, or 31P NMR experiments were carried out with a Bruker MSL 400, operating at 400 MHz (1H), 100.64 MHz (13C), or 161.9 MHz (31P). The coupling constants (J) are given in Hz. The direct inlet mass spectra were measured with a Finnigan MAT 90 (Bremen FRG), source temperature 150 °C (70 eV IP). The GCMS spectra were measured with a Hewlett Packard 5970 MSD equipped with a 5890 GC with 50 m CP Sil 5CB column. In the mass spectra, the appropriate isotope pattern was observed unless otherwise stated. Elemental analysis were carried out by Micro Analytisches Labor Pascher in Remagen, Germany. Melting points (uncorrected) were determined with melting point equipment of Pleuger after Dr. Tottoli. Dichloro(diiodomethyl)phosphine (2). A solution of isopropylmagnesium chloride, prepared from Mg (twice sublimed, 5.37 g, 221 mmol), isopropyl chloride (35 mL, 274 mmol), and THF (60 mL), was added dropwise with a syringe to a solution of iodoform (80.46 g, 204 mmol) in THF (400 mL) at -95 °C. The resulting dark red solution was frozen with liquid nitrogen to -196 °C. Then a Schlenk tube containing freshly distilled PCl3 (230 mL, 2.636 mol) was frozen with liquid nitrogen, too, and attached upside down upon the solid reaction mixture. By spontaneous warming, the PCl3 was allowed to flow down slowly on top of the carbenoid mixture which was kept at -196 °C. After complete addition, the reaction mixture was allowed to warm to -100 °C and mechanical stirring was commenced as soon as possible. The reaction mixture became yellow and a large amount of colorless precipitate was formed within 15 min. The reaction mixture was warmed to -40 °C during 4 h and then quickly to room temperature, while the color changed gradually from yellow to brown orange. The volatile compounds were removed under reduced pressure, and the residue was extracted three times with pentane (640 mL) and filtered. After concentration of the pentane extracts to 300 mL, another filtration was carried out. Further removal of pentane from the filtrate under vacuum furnished crude 2 as a red brown oil. Attempted purification via distillation or crystallization did not improve the purity significantly. Therefore, the subsequent Diels-Alder reactions with a 1,3-butadiene were carried out with freshly prepared crude 2, which, on the basis of 31P NMR spectroscopy was approximately 70% pure. The main impurity, δ(31P) 175 ppm, is presumably a

2-Iodophosphinines compound R-PCl2, but it was not identified. Data for 2 are as follows. NMR (C6D6): δ(1H) 4.34 (d, 2J(PH) ) 17.1); δ(13C{1H}) -19.1 (dd, 1J(PC) ) 84.9, {1J(CH) ) 167.5}); δ(31P{1H}) 155.2. HRMS (m/e): calcd for CH35Cl2I2P, 367.7280; found, 367.7256. MS (EI): m/e 368 ([M]+, 54), 333 ([M - Cl]+, 15), 241 ([M - I]+, 98), 206 ([M - 2Cl - I]+, 15). 2-Iodophosphinine (5a). Compound 2 was prepared as described above from iodoform (177.2 g, 450 mmol). In this case, commercial isopropylmagnesium chloride was used. Crude 2 was obtained in a yield of 63% (104 g, 0.28 mol). Then 1,3-butadiene (350 g, 6.5 mol) was condensed into the reaction vessel. After dropwise addition of approximately 1 equiv of NEt3 (39 mL, 0.28 mol) at -5 °C, a large amount of brownred precipitate was formed in the yellow solution. As this made stirring difficult, THF (270 mL) was added. Then the rest of the NEt3 (400 mL, 2.88 mol) was added dropwise within 3 h. Meanwhile, the temperature was allowed to rise to room temperature. After 16 h, the heterogeneous yellow brown mixture was evaporated and the residue extracted six times with pentane (800 mL). The combined extracts were concentrated to 100 mL and filtered. Distillation, first under normal pressure and finally under vacuum, furnished 5a (slightly contaminated with 1,1-diiodoisobutane) as a yellow liquid (bp 45 °C/4 × 10-3 mbar) in a yield of 10% (10.8 g, 0.044 mol). Data for 5a are as follows. NMR (C6D6): δ(1H) 6.61 (ddt, 4 J(PH) ) 4.3, 4J(HH6) ) 1.0, 3J(HH5) ) 8.4, 3J(HH3) ) 8.6, 1H, H4), 7.27 (apparent q, 3J(PH) ) 8.6, 3J(HH4) ) 8.4, 3J(HH6) ) 10.1, 1H, H5), 7.94 (ddd, 2J(PH) ) 38.8, 3J(HH5) ) 10.1, 4J(HH ) ) 1.0, 1H, H ), 7.95 (dd, 3J(PH) ) 5.3, 3J(HH ) ) 8.6, 4 6 4 1H, H3); δ(13C{1H}) 124.9 (dt, 1J(PC) ) 77.5, {3J(CH6) ) 12.1}, {3J(CH4) ) 12.1}, C2), 130.7 (dddd, 3J(PC) ) 18.6, {1J(CH) ) 159.6}, {3J(CH6) ) 8.8, 2J(CH3 or CH5) ) 2.0}, C4), 131.8 (ddd, 2J(PC) ) 13.3, {1J(CH) ) 158.3}, {3J(CH) ) 7.4}, C ), 143.4 5 (dddt, 2J(PC) ) 14.5, {1J(CH) ) 164.1}, {3J(CH5) ) 7.8}, {2J(CH4) ) 1.4}, {4J(CH6) ) 1.4}, C3), 159.4 (ddd, 1J(PC) ) 60.5, {1J(CH) ) 165.6}, {3J(CH4) ) 9.0}, C6); δ(31P{1H}) 233.1. MS (EI): m/e 222 ([M]+, 20), 95 ([M - I]+, 18). HRMS (m/e): calcd for C5H4IP, 221.9094; found, 221.907. Anal. Calcd for C5H4IP: C, 27.06; H, 1.82. Found: C, 26.94; H, 2.14. 2-Iodo-4,5-dimethylphosphinine (5b). To the crude mixture of 2, freshly distilled 2,3-dimethyl-1,3-butadiene (300 mL, 2.65 mol) was added. After cooling of the mixture to 0 °C, NEt3 (300 mL, 2.16 mol) was added dropwise during 5 h. A brown precipitate was formed, and the reaction mixture became red. The temperature was then raised gradually to room temperature. During 55 h the reaction was followed by 31P NMR spectroscopy, which indicated the conversion of 4b (δ(31P) 94 ppm) to 5b (impurity at (δ(31P) 95 ppm, 7%). After the mixture was stirred, the volatile fraction was removed under vacuum. The residue was extracted four times with pentane (600 mL) and filtered. After concentration of the combined pentane fractions to 150 mL and filtration, the filtrate was subjected to distillation, first at atmospheric pressure to remove pentane and then under vacuum to obtain 5b. A second distillation furnished pure 5b as yellow crystals (mp 30 °C; bp 76 °C/10-3 mbar) in a yield of 32% (16.4 g, 65.7 mmol) relative to iodoform. Data for 5b are as follows. NMR (C6D6): δ(1H) 1.69 (d, 5J(PH) ) 3.7, 3H, 4-CH3), 1.74 (s, 3H, 5-CH3), 7.69 (d, 2J(PH) ) 38.6, 1H, H6), 7.96 (d, 3J(PH) ) 5.6, 1H, H3); δ(13C{1H}) 21.5 (dq, 4J(PC) ) 1.9, {1J(CH) ) 127.2}, 1J(CC ) ) 43.7, 4-CH ), 22.7 (ddq, 3J(PC) ) 3.4, {3J(CH) ) 6.3}, 4 3 {1J(CH) ) 127.0}, 1J(CC5) ) 42.6, 5-CH3), 121.0 (dd, 1J(PC) ) 71.4, {3J(CH) ) 12.7}, 1J(CC3) ) 55.6, C2), 141.4 (d, 3J(PC) ) 17.5, 1J(CC3) ) 58.7, 1J(C4-CH3) ) 43.7, C4), 141.9 (d, 2J(PC) ) 14.8, 1J(C5-CH3) ) 42.6, 1J(CC6) ) 55.3, C5), 145.7 (ddq, 2J(PC) ) 15.2, {1J(CH) ) 165.8}, {3J(CH) ) 5.4}, 1J(CC4) ) 58.7, 1J(CC2) ) 55.6, C3), 159.7 (ddq, 1J(PC) ) 56.9, {1J(CH) ) 152.8}, {3J(CH) ) 5.0},1J(CC5) ) 55.3, C6); δ(31P{1H}) 216.2. HRMS (m/e): calcd for C7H8IP, 249.9407; found, 249.9446. MS (EI): m/e 250 ([M]+, 62), 123 ([M - I]+, 11). Anal. Calcd for C7H8IP: C, 33.63; H, 3.23; I, 50.76; P, 12.39. Found: C, 33.58; H, 3.20; I, 50.4; P, 12.6.

Organometallics, Vol. 15, No. 2, 1996 799 (η1-2-Iodophosphinine)pentacarbonyltungsten (7a). A solution of (acetonitrile)pentacarbonyltungsten (6) (7.2 g, 19.7 mmol) and 5a (4.02 g, 16.3 mmol) in THF (150 mL) was stirred at room temperature for 49 h. After 31 h, 0.44 g of 6 (1.2 mmol) was added to drive the complexation of 5a toward completion. Afterward, the dark green solution was evaporated and the residue was extracted with pentane (600 mL). Crystallization by cooling the extract to -70 °C furnished pure 7a as brown yellow crystals (dec 81 °C) in a yield of 48% (4.31 g, 7.9 mmol). Data for 7a are as follows. NMR (C6D6): δ(1H) 6.20 (tdd, 4J(PH) ) 8.0, 3J(HH5) ) 8.1, 3J(HH3) ) 8.8, 4J(HH6) ) 1.3, 1H, H4), 6.89 (dddd, 3J(PH) ) 22.7, 3J(HH6) ) 10.1, 3J(HH ) ) 8.1, 4J(HH ) ) 1.0, 1H, H ), 7.48 (ddd, 2J(PH) ) 4 3 5 26.4, 3J(HH5) ) 10.1, 4J(HH4) ) 1.3, 1H, H6), 7.68 (ddd, 3J(PH) ) 16.3, 3J(HH4) ) 8.8, 4J(HH5) ) 1.0, 1H, H3); δ(13C{1H}) 120.8 (dddt, 1J(PC) ) 2.0, {3J(CH4) ) 10.5}, {3J(CH6) ) 13.4}, {2J(CH2) ) 2.9}, {4J(CH5) ) 2.9}, C2), 127.6 (dddd, 3J(PC) ) 29.0, {1J(CH) ) 163.2}, {3J(CH6) ) 8.8}, {2J(CH3 or CH5) ) 1.5}, C4), 135.7 (ddddd, 2J(PC) ) 16.5, {1J(CH) ) 161.1}, {3J(CH3) ) 8.2}, {2J(CH6 or CH4) ) 1.2}, {2J(CH4 or CH6) ) 2.3}, C5), 146.2 (dddt, 2J(PC) ) 13.3, {1J(CH) ) 165.2}, {3J(CH5) ) 8.5}, {4J(CH6) ) 1.4}, {2J(CH4) ) 1.4}, C3), 153.0 (ddddd, 1J(PC) ) 12.4, {1J(CH) ) 162.5}, {3J(CH4) ) 9.1}, {2J(CH5) ) 2.1}, {4J(CH3) ) 1.5}, C6), 194.4 (d, 2J(PC) ) 9.3, 1 J(WC) ) 125.0, CO [cis]), 198.5 (d, 2J(PC) ) 33.0, CO [trans]); δ(31P{1H}) 201.5 (1J(PW) ) 285.6). HRMS calcd for C10H4O5IP182W, 543.8323; found, 543.832. MS (EI): m/e 546 ([M]+, 55), 518 ([M - CO]+, 3), 490 ([M - 2CO]+, 42), 462 ([M 3CO]+, 31), 434 ([M - 4CO]+, 34), 406 ([M - 5CO]+, 100). Anal. Calcd for C10H4IO5PW: C, 22.0; H, 0.74; P, 5.67; I, 23.25; W, 33.68. Found: C, 21.85; H, 0.76; P, 5.46; I, 22.0; W, 33.8. (η 1 -2-Iodo-4,5-dimethylphosphinine)pentacarbonyltungsten (7b). During 18 h, a THF solution (200 mL) of 7b (3.7 g, 14.8 mmol) and 6 (6.4 g, 17.5 mmol) was heated to 45-50 °C. After removal of THF under reduced pressure, the residue was extracted with pentane (200 mL) and crystallized by cooling the extract to -70 °C. Pure 7b was obtained as brown yellow crystals (dec 126-7 °C) in a yield of 86% (7.36 g, 12.8 mmol). Data for 7b are as follows. NMR (C6D6): δ(1H) 1.50 (s, 3H, 5-CH3), 1.54 (d, 5J(PH) ) 6.4, 3H, 4-CH3), 7.54 (d, 2J(PH) ) 25.6, 1H, H ), 7.73 (d, 3J(PH) ) 17.4, 1H, H ); δ(13C6 3 {1H}) 21.1 (ddq, 4J(PC) ) 4.0, {3J(CH) ) 4.0}, {1J(CH) ) 127.5}, 4-CH3), 22.4 (ddq, 3J(PC) ) 9.5, {3J(CH) ) 6.4}, {1J(CH) ) 128.0}, 5-CH3), 117.0 (t, 1J(PC) ) 6.6, {3J(CH) ) 6.6}, C2), 138.7 (d, 3J(PC) ) 26.2, C4), 146.7 (d, 2J(PC) ) 16.7, C5), 148.1 (ddq, 2J(PC) ) 13.3, {1J(CH) ) 161.9}, {3J(CH) ) 5.5}, C3), 152.5 (ddq, 1J(PC) ) 14.7, {1J(CH) ) 158.4}, {3J(CH) ) 5.2}, C6), 194.7 (d, 2J(PC) ) 9.3, 1J(WC) ) 125.1, CO [cis]), 198.7 (d, 2J(PC) ) 32.3, 1J(WC) ) 148.2, CO [trans]); δ(31P{1H}) 184.6 (1J(PW) ) 280.4). HRMS (m/e): calcd for C12H8IO5PW, 571.8636; found, 571.861. MS (EI): m/e 574 ([M]+, 32), 518 ([M - 2CO]+, 18), 490 ([M - 3CO]+, 62), 462 ([M 4CO]+, 20), 434 ([M - 5CO]+, 62). Anal. Calcd for C12H8IO5PW: C, 25.11; H, 1.41; I, 22.11; P, 5.40; W, 32.03. Found: C, 25.36; H, 1.47; I, 21.8; P, 5.46; W, 31.5. (η 1 -2-Iodo-4,5-dimethylphosphinine)pentacarbonylchromium (8). A mixture of di-n-butyl ether (50 mL), 5b (0.38 g, 1.5 mmol), and Cr(CO)6 (0.49 g, 2.2 mol) was heated at 140 °C for 5.5 h. A black precipitate was formed in the dark green solution. Removal of solvent under reduced pressure was followed by extraction of the residue with pentane (100 mL). Crystallization by cooling the extract to -70 °C furnished pure 8 as yellow crystals (dec 101 °C) in a yield of 12% (0.08 g, 0.18 mmol). Data for 8 are as follows. NMR (C6D6): δ(1H) 1.50 (s, 3H, 5-CH3), 1.55 (d, 5J(PH) ) 6.2, 3H, 4-CH3), 7.58 (d, 2J(PH) ) 26.2, 1H, H6), 7.70 (d, 3J(PH) ) 16.7, 1H, H3); δ(13C{1H}) 21.0 (ddq, 4J(PC) ) 3.9, {3J(CH) ) 4.5}, {1J(CH) ) 127.4}, 4-CH3), 22.5 (ddq, 3J(PC) ) 9.1, {3J(CH) ) 6.3}, {1J(CH) ) 127.8}, 5-CH3), 118.9 (d, 1J(PC) ) 2.5, C2), 138.6 (d, 3J(PC) ) 25.5, C4), 146.4 (d, 2J(PC) ) 17.0, C5), 148.8 (ddq, 2J(PC) ) 14.1, {1J(CH) ) 161.8}, {3J(CH) ) 5.4}, C3), 153.0 (ddq, 1J(PC) ) 8.8, {1J(CH) ) 157.8}, {3J(CH)

800

Organometallics, Vol. 15, No. 2, 1996

) 5.1}, C6), 214.9 (d, 2J(PC) ) 17.2, CO [cis]), 220.9 (d, 2J(PC) ) 3.8, CO [trans]); δ(31P{1H}) 230.9. MS (EI): m/e 442 ([M]+, 22), 414 ([M - CO]+, 3), 386 ([M - 2CO]+, 13), 358 ([M 3CO]+, 15), 330 ([M - 4CO]+, 31), 302 ([M - 5CO]+, 100), 250 ([M - Cr(CO)5]+, 11). Anal. Calcd for C12H8CrIO5P: C, 32.60; H, 1.82; I, 28.71; P, 7.01. Found: C, 32.36; H, 1.84; I, 27.10; P, 7.43. (η1-2-Lithiophosphinine)pentacarbonyltungsten (11a). A solution of 7a (0.15 g, 0.27 mmol) in THF (3.0 mL) was cooled to -100 °C. After addition of 1.6 M n-BuLi (0.25 mL, 0.40 mmol), the mixture was vigorously stirred while the temperature was maintained between -90 and -100 °C. The resulting black solution was subjected to low temperature 31P NMR measurements. NMR (THF) for 11: δ(31P{1H}) 176.9 (1J(PW) ) 235.2). (η 1 -2-(Triphenylstannyl)phosphinine)pentacarbonyltungsten (13a). A solution of 7a (1.12 g, 2.05 mmol) in THF (10 mL) was cooled to -100 °C. Then 1.6 M n-BuLi (1.3 mL, 2.08 mmol) was added during 30 s while the temperature was maintained at -100 °C. After 2 min, a solution of ClSnPh3 (0.90 g, 2.34 mmol) in THF (2 mL) was added during 10 s. Then the dark red reaction mixture was allowed to warm to room temperature. Evaporation of the reaction mixture under reduced pressure was followed by extraction of the residue with pentane (50 mL). Crystallization by cooling the extract to -70 °C furnished pure 13a as light brown crystals (dec 67 °C) in a yield of 4% (0.07 g, 0.09 mmol) relative to 7a. Data for 13a are as follows. NMR (C6D6): δ(1H) 6.70 (q, 4J(PH) ) 7.8, 3J(HH5) ) 7.8, 3J(HH3) ) 7.8, 1H, H4), 7.00-7.30 (m, 9H, meta,para-SnPh3), 7.40-7.60 (m, 7H, orthoSnPh3 + H5), 7.95 (dd, 3J(PH) ) 27.3, 3J(HH) ) 7.8, 1H, H3), 8.40 (dd, 2J(PH) ) 24.1, 3J(HH5) ) 10.2, 1H, H6); δ(13C) 126.5 (d, 3J(PC) ) 39.4, C4), 129.2 (s, 3J(119SnC) ) 54.6, 3J(117SnC) ) 52.5, SnPh3 meta-C), 129.9 (s, 4J(SnC) ) 11.9, SnPh3 paraC), 135.2 (d, 2J(PC) ) 31.8, C5), 137.5 (s, 2J(SnC) ) 39.8, SnPh3 ortho-C), 137.7 (d, 3J(PC) ) 2.7, 1J(119SnC) ) 553.1, 1J(117SnC) ) 528.7, SnPh3 ipso-C), 147.1 (d, 2J(PC) ) 20.7, 2J(SnC) ) 11.8, C3), 154.5 (d, 1J(PC) ) 7.9, 3J(SnC) ) 34.0, C6), 162.7 (d, 1J(PC) ) 30.2, C2), 195.4 (d, 2J(PC) ) 9.1, 1J(WC) ) 124.9, CO [cis]), 197.8 (d, 2J(PC) ) 29.2, CO [trans]); δ(31P{1H}) 195.4 (2J(P119Sn) ) 193.7, 2J(P117Sn) ) 185.5, 1J(PW) ) 269.7). HRMS (m/e): calcd for C28H19O5P116Sn184W, 765.9501; found, 765.949. MS (EI): m/e 768 ([M]+, 14), 684 ([M - 3CO]+, 32), 628 ([M 5CO]+, 42). Anal. Calcd for C28H19O5PSnW: C, 43.73; H, 2.49. Found: C, 43.54; H, 2.61. (η1-2-Lithio-4,5-dimethylphosphinine)pentacarbonyltungsten (11b). A solution of 7b (0.10 g, 0.17 mmol) in THF (2.5 mL) was cooled to -100 °C. After addition of 1.6 M n-BuLi (0.11 mL, 0.18 mmol), the mixture was vigorously stirred while the temperature was maintained between -90 and -100 °C. The resulting dark red solution was subjected to low-temperature 31P NMR measurements. NMR (THF) for 11b: δ(31P{1H}) 163.4 (1J(PW) ) 239.8). (η1-2-(Triphenylstannyl)-4,5-dimethylphosphinine)pentacarbonyltungsten (13b). To a cooled solution (-95 °C) of 7b (0.88 g, 1.53 mmol) in THF (10 mL) was added 1.6 M n-BuLi (0.95 mL, 1.52 mmol) during 10 min. After 20 min of stirring, ClSnPh3 (0.62 g, 1.61 mmol) was added to the dark red solution. After being stirred for 10 min at -90 °C, the reaction mixture was allowed to warm to room temperature. The solvents were removed from the resulting dark red solution, and the residue was extracted with pentane (50 mL). Crystallization by cooling the extract to -70 °C furnished pure 13b as yellow crystals (mp 144-5 °C) in a yield of 49% (0.60 g, 0.75 mmol) relative to 7b. Data for 13b are as follows. NMR (C6D6): δ(1H) 1.67 (d, 5J(PH) ) 6.2, 3H, 4-CH3), 1.77 (s, 3H, 5-CH3), 7.20-7.35 (m, 9H, meta,para-SnPh3), 7.60-7.75 (m, 6H, ortho-SnPh3), 7.86 (d, 3J(PH) ) 29.5, 3J(SnH) ) 66.6, 1H, H3), 8.36 (d, 2J(PH) ) 23.7, 4J(SnH) ) 23.8, 1H, H6); δ(13C{1H}) 21.6 (dq, 4J(PC) ) 4.0, {1J(CH) ) 127.5}, 4-CH3), 23.0 (dq, 3J(PC) ) 8.2, {1J(CH) ) 127.5}, 5-CH3), 129.2 (d, 3J(119SnC) ) 54.4, 3J(117SnC) ) 52.4, {1J(CH) ) 157.3}, SnPh3 meta-C),

Teunissen and Bickelhaupt 129.8 (dt, 4J(SnC) ) 11.7, {1J(CH) ) 159.6}, {3J(CH) ) 7.0}, SnPh3 para-C), 137.5 (d, 2J(SnC) ) 38.4, {1J(CH) ) 157.0}, SnPh3 ortho-C), 136.9 (d, 3J(PC) ) 35.4, C4), 137.9 (d, 3J(PC) ) 2.7, 1J(119SnC) ) 551.0, 1J(117SnC) ) 526.5, SnPh3 ipso-C), 148.5 (d, 2J(PC) ) 15.5, C5) 149.2 (ddq, 2J(PC) ) 20.8, {1J(CH) ) 157.1}, {3J(CH) ) 5.4}, C3), 154.1 (ddq, 1J(PC) ) 10.4, 3J(SnC) ) 37.6, {1J(CH) ) 158.4}, {3J(CH) ) 5.4}, C ), 159.0 6 (d, 1J(PC) ) 25.5, 1J(119SnC) ) 404.6, 1J(117SnC) ) 385.5, C2), 195.7 (d, 2J(PC) ) 9.2, 1J(WC) ) 124.5, CO [cis]), 198.0 (d, 2 J(PC) ) 28.7, 1J(WC) ) 145.9, CO [trans]); δ(31P{1H}) 180.5 (2J(P117Sn) ) 186.9, 2J(P119Sn) ) 195.8, 1J(PW) ) 266.2); δ(119Sn) -107.6. HRMS (m/e): calcd for C30H23O5P120Sn182W, 795.9814; found, 795.984. MS (EI): m/e 798 ([M]+, 28), 770 ([M - CO]+, 6), 712 ([M - 3CO]+, 42), 656 ([M - 5CO]+, 64), 579 ([M - 5CO - Ph]+, 5). Anal. Calcd for C30H23O5PSnW: C, 45.22; H, 2.89. Found: C, 45.28; H, 3.29. 2-(Iodomagnesio)-4,5-dimethylphosphinine (17) and 4,4′,5,5′-Tetramethyl-2,2′-biphosphinine (16). (a) In the Absence of Triphenylphosphine. To a mixture consisting of Mg (three times sublimed, 0.10 g, 4.11 mmol) and THF (5 mL) was added 1,2-dibromoethane (0.2 mL, 2.32 mmol). After the exothermal activation of Mg, the mixture was cooled to room temperature and 5b (0.29 g, 1.16 mmol) was added during stirring. The reaction mixture became black, and a dark brown precipitate was formed. After 25 h of stirring at room temperature, 31P NMR spectroscopy indicated the presence of 17, 16, and 18 in a ratio of 19.2:3.6:18.4. (b) In the Presence of Triphenylphosphine. Reliable indications concerning the yield of the products and the degree of decomposition of 5b were obtained by carrying out a reaction with triphenylphosphine as internal standard, assuming that triphenylphosphine is inert under the reaction conditions. Therefore, the relative molar ratio of 5b and triphenylphosphine was related to their relative intensity in a 31P NMR spectrum, recorded before the reaction of 5b with magnesium. An intensity ratio of 3.22:1 (5b:triphenylphosphine) corresponded to a molar ratio of 1:1. This ratio, deduced from 5b, was applied to all phosphinine derivatives formed later. In this experiment, magnesium (6.17 mmol) was activated with 1,2-dibromoethane (4.48 mmol) in THF (10 mL) in the presence of triphenylphosphine (0.95 mmol). After cooling of the mixture to room temperature, 5b (1.12 mmol) was added after which the reaction was followed by 31P NMR spectroscopy. After 6 days of stirring at room temperature, the black reaction mixture, in which a dark brown precipitate was present, contained 18, 16, and 17 in an absolute yield of 41.5, 1.3, and 19.2%. More than one-third (37%) of 5b had decomposed. Another 6 days later, only 0.11 mmol (10%) of phosphinines was left: predominantly 18 and hardly any 16 and 17. Data for 17 are as follows. NMR (THF): δ(31P{1H}) 229.0 ({2J(PH) ) 3J(PH) ) 26.7}). Data for 16 are as follows. NMR (C6D6): δ(31P{1H}) 181.3. HRMS (m/e): calcd for C14H16P2, 246.0727; found, 246.0752. MS (EI): m/e 246 ([M]+, 100), 231 ([M CH3]+, 33). (4,5-Dimethylphosphininyl-κC2)(iodo)(TMEDA)zinc (20). A reaction mixture consisting of 5b (1.88 g, 7.52 mmol), TMEDA (2.29 g, 19.7 mmol), and zinc powder (2.86 g, 43.7 mmol) in THF (12 mL) was stirred for 19 h at room temperature. Approximately 8 mL of this solution was cooled to -70 °C. After several days, a precipitate was formed consisting of dark gray crystals, contaminated with some light brown powder and a small amount of zinc dust. After removal of supernatant, the precipitate was dried and subjected to NMR analysis. Data for 20 are as follows. NMR (C6D6): δ(1H) 1.88 (s, 4H, NCH2), 2.15 (s, 12H, NCH3), 2.20 (s, 3H, 5-CH3), 2.24 (d, 5J(PH) ) 3.4, 3H, 4-CH3), 8.67 (d, 2J(PH) ) 31.9, 1H, H6), 8.81 (d, 3J(PH) ) 18.7, 1H, H3), ratio TMEDA/20 1:1; δ(13C) 22.4 (d, 3J(PC) ) 1.8, 5-CH3), 23.6 (d, 4J(PC) ) 1.8, 4-CH3), 49.0 (s, NCH3), 56.8 (s, NCH2), 136.3 (d, 3J(PC) ) 29.8, C4), 141.5 (d, 2J(PC) ) 15.1, C3), 147.9 (d, 2J(PC) ) 16.3, C5), 155.0 (d, 1J(PC) ) 63.2, C6), 183.4 (d, 1J(PC) ) 86.6, C2); δ(31P{1H}) 220.5.

2-Iodophosphinines 2-(Iodozincio)-4,5-dimethylphosphinine (21). A mixture containing 5b (0.18 g, 0.7 mmol), zinc powder (0.19 g, 2.8 mmol), and DMF-d7 (1 mL) was stirred at room temperature for 15 h. Then the dark red solution was separated from the excess of zinc powder and placed in an NMR tube which was subsequently sealed. Data for 21 are as follows. NMR (DMFd7): δ(1H) 2.32 (d, 5J(PH) ) 3.3, 3H, 4-CH3), 2.35 (s, 3H, 5-CH3), 8.14 (d, 3J(PH) ) 18.8, 1H, H3), 8.42 (d, 2J(PH) ) 31.6, 1H, H6); δ(13C{1H}) 20.7 (dddq, 4J(PC) ) 1.7,{3J(CH) ) 6.0}, {4J(CH) ) 0.9}, {1J(CH) ) 125.6}, 4-CH3), 20.9 (dddq, 3J(PC) ) 1.4, {3J(CH) ) 6.8}, {4J(CH) ) 1.4}, {1J(CH) ) 126.0}, 5-CH3), 133.8 (d, 3J(PC) ) 29.7, C4), 139.6 (d, 2J(PC) ) 14.7, C5), 143.8 (ddq, 2J(PC) ) 16.5), {1J(CH) ) 152.9}, {3J(CH) ) 5.1}, C3), 154.2 (ddq, 1J(PC) ) 64.0, {1J(CH) ) 149.0}, {3J(CH) ) 5.1}, C6), 184.4 (d, 1J(PC) ) 86.6, C2); δ(31P{1H}) 222.6. [(4,5-Dimethylphosphininyl-1KP,2KC 2 )pentacarbonyltungsten](iodo)(TMEDA)zinc (22). A solution of 20 (3.16 mmol) in THF (6.2 mL) was added to a solution of 6 (1.17 g, 3.21 mmol) in THF (2 mL). After 23 h of stirring at room temperature, a small amount of yellow precipitate was formed

Organometallics, Vol. 15, No. 2, 1996 801 in the dark red solution. Of this solution a small amount (2 mL) was evaporated under reduced pressure. Subsequent NMR measurements indicated complete decomposition after 7 days at room temperature. Data for 22 are as follows. NMR (THF-d8): δ(1H) 2.22 (s, 3H, 5-CH3), 2.52 (d, 5J(PH) ) 9.9, 3H, 4-CH3), 2.49 (s, 12H, NCH3), 2.68 (s, 4H, NCH2), 8.16 (d, 3J(PH) ) 40.4, 1H, H3), 8.26 (d, 2J(PH) ) 21.2, 1H, H6); δ(13C) 21.9 (d, 4J(PC) ) 4.2, 4-CH3), 22.8 (d, 3J(PC) ) 6.9, 5-CH3), 49.2 (s, NCH3), 57.3 (s, NCH2), 134.5 (d, 3J(PC) ) 42.1, C4), 145,4 (d, 2 J(PC) ) 16.5, C5), 149.0 (d, 2J(PC) ) 24.1, C3), 154.3 (s, C6), 180.2 (d, 1J(PC) ) 45.3, C2), 197.6 (d, 2J(PC) ) 9.5, CO [cis]), 200.8 (d, 2J(PC) ) 24.1, CO [trans]); δ(31P{1H}) 174.6 (1J(PW) ) 249.7).

Acknowledgment. We are grateful to Dr. B. L. M. van Baar for measuring the HRMS spectra and to Dr. F. J. J. de Kanter for the NOE and INADEQUATE experiments. OM9505322